_L.-

,7

"'---'-

. JPL/PUBLICATION 84-16-'~

NASA-CR-17372819840019646

\- ,

S,ateUite. En,~,ssi'orr:Ra'nge ,Inferred," .,Earth, Surlvey ($ERIE.S)- ProJect:, Final

.~.Rep,ort on· R.es~arch.,and Development', ,I,

Phase~',1,979to"1~83 .' ~/ .

" L.A. Buennagel" ·P.F. MacDoran ', .' R~E. Neilan

. D.J. Spitzmesser:LE~ Young-

/

/

" ,

~ - / rj..<

, March 1, 1984.

;

"

',_NI\Sl\. 'National Aeron~utic~a'nd',' "­

'Space Administration]

Jet· Propulsion Laboratory <

/ \ ---:- 'California'-Inst'itute of--Tecnnology,Pasadena; California ' -

"- -- /

,/

, ,,

0--.....i,'d·V?, ') 199L1

_1. ..... ~_.:}, ~L.-. ',~,

'.,:, LANGLEY RESEARCH CENTER

L18RARY.N~SI\ ,',

. ~ HA.~.1~TON,.VJRGINIA :'

: --"

/,

.' .--1

''-----,'

https://ntrs.nasa.gov/search.jsp?R=19840019646 2018-07-11T10:57:46+00:00Z

~":lPur;~nL-tH i l!J.:i-.'

".A T .-''' T '.1,,-'iii:". i '-.'-. i i',il_I J J. '_' __' J. I "!~_-:

================================================================================.1 1 RN/JPL-PUB-84-16

f""'.T,....,...,,1 , ..... ~r- I·--. ... Aiii ·-.. ..... i 'v' _-':-~ ,/ .,r"'; ;L.'. J. __'I !-i i I L..._•••• .::.- ...... .(I

34N27714** ISSUE 18 ~PAGE 2791 CATEGORY 4;.)PL-PtJB-E>i-15 f\lAS 1~',26~ 173728 Cr:·,lT=ti=: f:':lAS7-9·1E:'11'.'.-1 .f~r--.. T r- T r-r--. r"'.r-...~I ....... ..-P.• ..,..

l.if'Jl-LH::J:=i 1 r- 1 t.L} L.it)L·tjj'iit.i\i. t

RPT~= NASA-CR-173728

UTTL: Satellite Em1ss1on Range .Inferred Earth SurveY (SERIES) projectFinal Report, 1979 -. 1983

"T"I {,,""-M_

j L~r-';

MAJS: /*GEODESY/*GLOBAL POSITIONING SYSTEM/*MICROWAVE FREQUENCIES;*NAVIGATION/*'C"""'r...... /""-r-r-n/.. r-""!"'" "'T"r-1 ........~I/T ... I ...-·

~~HL~LNH~j j~HLKl~~~

MINS: / ACCURACY/ COST EFFECTIVENESS/ GUIDANCE CMOTION)/ RADIO STARS; VERY LONGBASE INTERFEROMETRY

AES; The Global Posittonjng SYstem (GPS) was developed by the Department ofr--__ ,-_. ._._-= ._'='1 •• r_._ .__ .. .=_._.J..:_._ •. __ 1- •• ~!__ !1._:.1._..I i"""-..L_~ __ .fo••

4! r- __ ~ __ ,_

L!eTefit:;e f-ir;Hictrt ~::f T()r [id\tlJ::.idlii}fi ~)t:;e ~)Y' tfle \...'f}i leu ~tdtet:; HfWeCj rt..:(c:et:;=

The sYstem will consist of a constel lat10u of 18 operational NavigationSatel1ite Tjmlng and Rangjng (NAVSTAR) satel1jtes by the late 1380!s:Dur~ng the last four Yea;s~ the Satel1 lt0 EmissjoD Range Inferred EarthSurveying (SERIES) team at the Jet Propulsion Laboratory (JPL) hasdeveloped a 00veJ receiver which is the heart of the SERIESsystem desjgned to use SfgnaJs broadcast from the GPS~ Th1s rece1ver does

JPL PUBLICATION 84-16

Satellite Emission Range InferredEarth Survey (SERIES) Project: FinalReport on Research and DevelopmentPhase, 1979 to 1983

L.A. BuennagelP.F. MacDoranR.E. NeilanD.J. SpitzmesserL.E. Young

March 1, 1984

NI\SI\National Aeronautics andSpace Administration

Jet Propulsion LaboratoryCalifornia Institute of TechnologyPasadena, California

The research described in this publication was carried out by the Jet PropulsionLaboratory, California Institute of Technology, under a contract with the NationalAeronautics and Space Administration.

Reference herein to any specific commercial product, process, or service by tradename, trademark, manufacturer, or otherwise, does not constitute or imply itsendorsement by the United States Government or the Jet Propulsion Laboratory,California Institute of Technology.

PREFACE

The work described in this report is a consequence of many years of research

and development activity on the use of radio interferometry at microwave

frequencies using compact radio stars as the sources of radio signals. The

initial SERIES engineering configuration, using the NAVSTAR satellites instead

of quasars, was to process the data by the Very Long Baseline Interferometry

(VLBI) techniques. This approach resulted in a complexity of data processing

which was not compatible with the concept of using the much stronger radio

signals. Fortunately, it turned out that using the one-way range measurement

from the satellite to each of the two earth stations greatly simplified the

entire project. Although the primary dedication of the research was to create

a system competitive in accuracy with the VLBI and laser ranging techniques

(but at much less cost in capital expenditures), two additional significant

results were achieved: 1) the ability to synchronize clocks to the few

nanosecond level at intercontinental distances in near real time, and 2) the

ability to measure the total columnar content of free electrons between the

NAVSTAR satellite and the earth receiver along the line of sight between them.

N. A. Renzetti, Manager

JPL Geodynamics Program

iii

ACKNOWLEDGEMENT

The authors wish to acknowledge the support of Mr. T. L. Fischetti

of NASA Headquarters for his foresight in funding the feasibility and

proof-of-concept work at JPL with regard to the application of the Global

Positioning System for precision earth surveying.

The authors gratefully acknowledge the assistance of their JPL

colleagues: M. G. Newsted, M. A. Smith and S. Harmon for the mechanical

design and fabrication of the portable stations; J. B. Thomas for theoretical

studies and system analysis; G. M. Resch and D. W. Curkendall for discussions

on system design options; H. N. Royden, R. M. Miller, and T. E. Litwin for VHF

Faraday rotation ionospheric monitoring at Goldstone, CA; A. L. Berkin for

assistance in taking data; S. J. DiNardo and C. J. Vegos for logistical

support. The cooperation of F. B. Varnum, Defense Mapping Agency and Sergeant

Wagner, Vandenberg AFB, is also acknowledged for making available the

approximate GPS satellite ephemerides used in the SERIES microcomputer for

pointing the 1.5 m antennas.

iv

CONTENTS

I. INTRODUCTION . · · · · · · . · · · · · · · · · · · · 1

II. STATEMENT OF THE CHALLENGE 5

III THEORY OF OPERATION · · · · · 6

IV. IMPLEMENTATION · · · · · · · · · 16

V. DIFFERENTIAL OBSERVATIONS · · · · · 22

VI. IONOSPHERIC CALIBRATIONS · · · · · · · · 23

VII. BASELINE ESTIMATION SOFTWARE · · · · · · · · · · · · · · 28

VIII. SERIES TEST SEQUENCE · · · · · · · · 29

IX. CURRENT SERIES STATUS · · · · 41

X. SYMBOL GLOSSARY · · · · · · · 43

REFERENCES . . . . . . . . · · · · · · . · · · · · · · · · · · · · · 45

Figures

1. One of the two SERIES proof-of-concept stations • • • •• 2

2. A two-dimensional illustration of conventional positioningusing a GPS receiver having code access • 7

3. SERIES differential positioning geometry 11

4.

5.

6.

7.

SERIES Doppler positioning Geometry

SERIES simplified block diagram • • •

SERIES delay and multiply technique ••

A schematic representation of the SERIES baselinesolution • • • • • • • • • • • • • • •

· .14

17

18

20

8.

9.

10.

SERIES technique for measurement of the columnar chargedparticle content • • • • • • • • •••

SERIES 150 m baseline length measurements

Comparison of SERIES and composite measurements of the21 km baseline •. . . . . • • . . . . • • . • • • • . .

v

25

31

33

11.

12.

13.

CONTENTS (Contd)

One nanosecond (10-9 second) agreement was observed oneach of two experiments in which a traveling clock wasused to verify SERIES clock synchronization measurementsmade on the 21 km baseline • • • • • • • • • • •

The P-codes are nearly in phase when transmitted fromthe GPS satellites, but suffer a differential delay dueto the presence of free electrons in the ionosphere • • •

The SERIES ionospheric measurements are shown as circles,while the Faraday rotation results are triangles

35

36

37

14. The calculated cumulative error in baseline length isshown for the SERIES receivers as tested •••••• . . 39

15. Table of SERIES baseline results

vi

. . . . . 40

ABSTRACT

The Global Positioning System (GPS) was developed by the Department of

Defense primarily for navigation use by the United States Armed Forces. The

system will consist of a constellation of 18 operational Navigation Satellite

Timing and Ranging (NAVSTAR) satellites by the late 1980's.

During the last four years, the Satellite Emission Range Inferred Earth

Surveying (SERIES) team at the Jet Propulsion Laboratory (JPL) has developed a

novel receiver which is the heart of the SERIES geodetic system designed to

use signals broadcast from the GPS. This receiver does not require knowledge

of the exact code sequence being transmitted. In addition, when two SERIES

receivers are used differentially to determine a baseline, few em accuracies

can be obtained.

The initial engineering test phase has been completed for the SERIES

Project. Baseline lengths, ranging from 150 meters to 171 kilometers, have

been measured with 0.3 em to 7 em accuracies. This technology, which is

sponsored by the NASA Geodynamics Program, has been developed at JPL to meet

the challenge for high precision, cost-effective geodesy, and to complement

the mobile Very Long Baseline Interferometry (VLBI) system for Earth surveying.

vii

I. INTRODUCTION

The commonly accepted model of the structure of the Earth's crust is that

it is composed of a set of tectonic plates. One of the stated objectives of

the NASA Geodynamics Program is to contribute to the understanding of the

processes that result in the movement and deformation of these tectonic

plates. The average rates of interplate motion over geologic time scales have

been determined from the geologic record to be in the range of a few

centitneters per year, but whether this motion is constant or episodic is

currently unknown.

Two space-based techniques, currently being supported by the Geodynamics

program, have been developed, and are capable, in principle, of measuring the

motions between plates in real time (a few years). These techniques are Very

Long Baseline Interferometry (VLBI) and Satellite LASER Ranging (SLR).

In addition to the large scale movement between tectonic plates, more

complicated motion in the Earth's crust is known to occur. An example of this

would be the local crustal distortion caused by the strain accumulation and

release near the boundaries of plates. It is necessary to describe such

motion in detail in order to be able to deduce the physical processes at work,

or to generalize such plate boundary interactions for study of the forces that

drive and resist plate motion. In order to adequately characterize these

distortions, it is necessary to measure a grid of reference points that is

densely sampled in space and time.

It is not possible to make such measurements using VLBI or SLk techniques

due to economic considerations. The requirement for the dense set of accurate

geodetic measurements, needed for regional geodesy and a number of related

studies, is the motivation for developing the Satellite Emission ~ange Inferred

Earth Surveying (SERIES) Global Positioning System (GPS) receiver which is

described in this report.

Two proof-of-concept SERIES stations (Fig. 1) have been constructed.

lhese stations each have a 1.S-m diameter prime feed parabolic dish antenna

1

N

Figure 1. One of the two SERIES proof-of-concept stations. Data aquisitionelectronics occupy a small portion of the camper shell. Theantenna is a 1.5 meter diameter, prime focus feed dish mountedon a two axis (azimuth and elevation) drive.

which provides approximately 23 dB of gain in the GPS signal. The receivers

have been constructed from off-the-shelf components wherever possible (Ref. 1).

The receivers contain a locally generated test signal for self-calibration.

Data can be recorded on either floppy disks or 9-track tapes.

The original SERIES implementation was to use the NAVSTARs as Radio

Frequency (RF) noise sources for Very Long Baseline Interferometry (VLBI)

terminals (Ref. 2). However, it was realized that the periodic structure in

the GPS signal could be used to provide ranging information even without code

sequence knowledge. By exploiting the periodic structure, the GPS signal

bandwidths were collapsed, and the data acquisition and post-processing were

greatly simplified.

The proof-of-concept statiohs have been tested on baselines up to 171 km

during the last year and a half. Baseline vectors determined with the SERIES

stations have demonstrated few em measurement accuracies for baselines of a

few tens of km (Ref. 3).

These SERIES stations are also capable of measuring the total columnar

content of free electrons integrated along the path between the receiver and

the GPS satellite. This measurement was validated through comparison with a

co-located Faraday rotation device, obtaining agreement to 1016 electrons

per meter squared (Ref. 4).

An additional feature of the SERIES technique is the ability to

synchronize clocks at different sites by receiving common signals from the

satellite. This was demonstrated by comparisons of clock offsets measured

with the SERIES stations against independent measurements made simultaneously

with an atomic clock carried between the two stations (Ref. 5). The agreement

between the two techniques was within 1 nanosecond (10- 9 second) for each of

the two experiments performed. The SERIES technique for clock synchronization

is expected to give accuracies of 1 to 10 nanoseconds for intercontinental

station separations.

3

Development of the SERIES technology is currently underway at JPL for the

following applications:

1. High-precision, cost-effective Earth surveying

2. Precision orbit determination of low-Earth orbiters

3. The precise positioning of marine platforms to aid in themeasurement of sea floor motions.

4

II. STATEMENT OF THE CHALLENGE

The satellite constellation of the GPS t when fully implemented in the

late 1980'st will be composed of 18 Earth satellites (Ref. 6). This constel­

lation will illuminate the Earth with dual L-band microwave signals intended

to serve t primarilYt U.S. military needs for positioning and navigation. At

this writing t six satellites are in orbit and functional. The three main

elements of the GPS are the satellite segment (termed NAVSTAR) which develops

L-band signals from on-board atomic clocks and disseminates satellite

ephemeris information; a ground control segment that monitors and controls the

ephemeris of each satellite and the clocks in orbit; and a user segment that

receives the L-band signals to accomplish a wide variety of positioning and

navigation functions. Because of the nature of military missions t each

segment of the GPS has capability which is restricted t and the highest preci­

sion NAVSTAR transmissions are encrypted in a manner originally designed to be

unavailable to the public.

In its highest precision mode for point positioning of a single receiver t

the GPS is required to have an accuracy of 16 meters in three dimensions.

However t for certain applications t such as crustal deformation monitoring to

provide inputs for geophysical modeling, sub-decimeter relative positioning

accuracy is the threshold of interest. The challenge is to devise a system

which (1) exploits the technology of the existing Global Positioning System;

(2) operates~ knowledge of the codes, while simultaneously obtaining an

accuracy level three orders of magnitude better than required by the original

GPS design; and (3) achieves these results with inexpensive equipment and

cost-effective data processing.

5

III. THEORY OF OPERATION

In order to understand the SERIES concept of GPS, it is useful to

consider the convential mode of operation in which the user has access to

the codes (see Fig. 2). Suppose an observer is located at a geocentric

vector position described as ~1 with the NAVSTAR-GPS located at geocentric+

vector position p. The observer acquires the satellite broadcast whens

the satellite rises above the horizon. Imagine that the clock aboard the

satellite, registering t , and the clock at the observer's radio receiver,s

registering t u ' are synchronized. If the apparent delay in the arrival of

the satellite clock pulses is ~t, then the apparent range to the satellite

is p = c ~t, where c is the speed of light.

The magnitude of the range varies between 25xl06 meters at satellite

rise and is set to a minimum of 20xl06 meters at the Time of Closest Approach

(TCA) to the observer's position. The rate of change of range causes the

P-code modulation to vary from its true frequency (f = 10.23 MHz) by an amount

equal to ±27 Hz with the true frequency value being observed at approximately

TCA. However, in the real world, neither the user's nor the satellite's clock

is exactly synchronized to an external frequency standard such as Universal

Time Coordinated (UTC). This results in an error in the observed range. The

observed range, p', now appears as

p' =1 p - R I + c(~t - ~t )s u u s

where the subscripts u and s denote user and satellite, respectively, and the

~t and ~t indicate clock bias with respect to the UTC. Also, theu s

measured frequency of the satellite signals, f, is in error by a summationmof errors in the clock rate of both user and satellite clocks, as well as

errors due to atmospheric range rate terms. This type of GPS measurement is

also known as single point or absolute positioning.

+In the conventional GPS configuration, the terms Ps and ~ts are

supplied to the user by a 50 bit-per-second telemetry message which contain~,

satellite ephemeris and clock state information. When a user decrypts

6

Pi = PSEUDO RANGE TO ith SATELLITE

USER RANGE ERRORDUE TO CLOCK OFFSETS: c(.~tu - ~t5i)

TRUE RANGE' p. = p~ - c(At - At .). I I U 51

NAVSTAR·GPS

1~(Xl' V,)

~t~TRUE USERPOSITION

APPARENTPOSITION

SATELLITECLOCK OFFSET

= USER CLOCK OFFSET

GIVEN INFO: Xi' Y i & Atsi

WHERE i = 1,20R3

SOLVE FOR: Xu- Yu & Atu

Figure 2. A two-dimensional illustration of conventional positioning using a GPSreceiver having code access

..

R ,u

as well as a fourth term which accounts for the user clock offset (Ref. 7).

signals from four NAVSTA.K satellites using the coarse acquisition (CiA) code

and the preclslon (P) code channels, the resulting measured pseudo ranges P'i'

permit solution of the user geocentric location, Ru • Four satellite

observations are necessary to solve for the three dimensional location,

The development of the SERIES system as described herein assumes no

knowledge of the CiA or P-code modulation sequence but can deduce differen­

tial positioning information, implementing a theory very different from the

design concept of conventional GPS use.

The first challenge is to extract a usable signal from the encoded L-band

transmissions without a decrypting receiver. Briefly, that detection is

accomplished by a delay and multiply process that recovers the chipping rates

of the CiA and P-code modulations on the dual L-band carriers, as described in

Section IV of this report. This type of information acquisition has a range

ambiguity, unlike the pseudo-range observable derived by decrypting the GPS

P-code signals. Having no dependence upon the coded form of GPS transmissions

is such an advantage to a civilian user community that innovations to accom-

TIlodate this ambiguous data are well worth the effort. This is particularly

attractive when the resultant system can achieve a differential accuracy

improvement 1000 times better than the 16m accuracy designed into the conven­

tional CPS mode. This improved accuracy is also possible for properly

designed conventional code correlating GPS receivers when used in a

differential mode.

The pseudo-range ambiguities are 2~.3m from P-code modulations extracted

from either the L1 or L2 channel and 2~3m from the CiA code modulation

which is available only on the L1 channel. The P-code ambiguity is resolved

by interpolating within the CiA code modulation to a precision of 10 meters

which represents a modest 3% of the CiA code cycle wavelength and requires an

amplitude Signal to Noise Ratio, SNR = 5. An ambiguity of some number of CiA

cycles (wavelength, 2~3m) is now present which can be resolved by the use of

Doppler positioning techniques accurate to 100 meters which are unambiguous.

Alternately, the 2~3 m ambiguities can be resolved using ~priori baseline

information. Thus, a method has been outlined which deduces unambiguous

pseuao ranges based on P-code modulation.

8

Consider the two dimensional case shown in Figure 2. To deduce the true

user position in absolute mode, a solution of the following system of

equations is performed:

~(Xu X )2 + (Y Y )2 pl1 - c(~t - 6t )1 u 1 u sl

v1(~ X )2 + (Y Y )2 p/2 - c(~t - ~t )2 u 2 u s2

~(~ _ X )2 + (Y - Y )2 p/3 - c(~tu - ~ts3)3 u 3

where p' l' p' 2' and p' 3 indicate actual measurements of the pseudo-range

from the user's receiver to the satellites which are located at the given

satellite positions of X. and Y. and with their individual satellite clock1 1

time offsets of 6t si where i = 1, 2 or 3.

For this two-dimensional case the three unknowns, Xu' Yu and ~tu are solved

for based upon the three observations P'1' P/2 and P/3. These measured

values are subsequently combined with corrections for various time delays to

produce an estimate of the true range, p.

The SERIES development, to be expanded upon in later sections, makes it

possible to measure the pseudo-ranges p'. without code knowledge. However,1

the satellite clock offset terms, 6t ., are generally unknown for each ofSl

the satellites being observed by a user without code knowledge. To avoid the

necessity of knowing satellite time, t s ' or the satellite clock offset,

6t si ' a pair of user stations now observes the three satellites, and the

measured pseudo ranges to each satellite are differenced between stations:

9

AllY offset in the epoch time of satellite transmission is an error presented

simultaneously to the two stations and is therefore common-mode eliminated in

the differential data type (see Fig. 3). This two-station differential data

type is also identical to the delay-data type developed by cross correlation

signal detection in Very Long Baseline Interferometry (VLBI).

lhus, the point positioning technique producing the geocentric location

of the user and the user clock offset (X , Y , ~t ) is abandoned inu u ufavor of a tecnnique for determining the differential location of station 2

A

relative to station 1. The unit vector S is located at station 1 and points

to the satellite signal source at a distance Pl' The baseline vector Bpoints from the antenna at station 2 to the antenna at station 1.

The satellite clock offsets are eliminated since they are common to each

station,Dut in their place a new clock parameter appears, that of the two

user clocks' offset. In general, clock 1 is not synchronized to clock 2.

Because the satellites are a finite distance away from the user receivers,

incomingwavefronts are spherical in shape. The VLBI geodetic experience

deals with extragalactic radio sources, which are virtually at infinity and

thus have planar wavefronts. As shown in Reference 8, the usual VLBI geodetic

equations can be recast into a modified form by including a spherical to

planar wavefront term, O. The set of differential pseudo-range equations

take on the form

(1)

where

and

~p' p' - P'2 1 (2)

where the prime indicates a measured quantity.

The term (&t u1 - &tu2 ) is the offset of clock 2 relative to clock 1 and

t TM is the differential time delay associated with the transmission media of

10

SPHERICALWAVEFRONT

WAVELENGTHAMBIGUITYRESOLVED BYDOPPLERPOSITIONING

~p= P2- Pl ~

~p' = -T (f2 T2 - f 1 T 1) + ~ (N2 - N1)

/\ I A-":>u.p = - S . B - {) + c(..1tu 1 -Atu2) + ctTM

~RA~SMI~IONI MEDIA ERROR

USER CLOCKDIFFERENTIALERROR

FREQ f 1COUNTER

STOPTIME T1INTERVAL

START COUNTER

Figure 3. SERIES differential positioning geometry

the troposphere and ionosphere through which the satellite signals must pass

in reaching the user receivers. The tropospheric delay calibration can be

ignored, modeled, or explicitly calibrated depending upon accuracy require­

ments. The ionospheric delay can be explicitly measured without code

knowledge by exploiting the dual frequency NAVSTAR transmissions, as will be

discussed in subsequent sections. The terms N1 and N2 account for the

phase ambiguities associated with the SERIES technique at stations 1 and 2.

For example, the P-code chipping rate of f = 10.23 MHz is recovered and

exhibits a phase offset indicative of the pseudo-range, P., to the1

satellite but with an ambiguity of the P-code wavelength. In Figure 3, the

pseudo range at station 1 is formed by measuring the elapsed time from the

one-pulse-per second (1 pps) event markers of clock 1 to the next occurring

positive-going-zero crossing of the P-code's 10.23 MHz sinusoidal output

encountered by receiver 1.

The time interval counter (TIC) indicates a value T1 between 0 and

97.75 nanoseconds (1/10.23 MHz). The received P-code frequency will be

Doppler shifted by up to 27 Hz for a stationary receiver, and the actual

received frequency, f 1, is measured. The P-code fractional phase is now

formed by multiplying the f 1 by T1• The pseudo range is the wavelength

(elf) multiplied by the fractional phase,

(3)

where N1 clf is the integer number of whole P-code wavelength cycles (elf =A = 29.3 m) between the receiver and satellite, and f is the rest frequency

of the satellite P-code clock.

In this process it has been assumed that the geocentric location ofA

receiver 1 (R) the unit vectors from receiver 1 to the satellite, (S), and

the range from receiver 1 to the satellite (Pi) are known quantities.

Receiver 2 outputs are now measured relative to clock 2 and the pseudo

range from station 2 becomes

12

The differential pseudo range is now formed as

(4)

!;,p , p'z - P'l (5)

The true differential pseudo range, ~p, is functionally dependent upon the

baseline as indicated in equation (1).

The problem is now the determination of the ambiguity of whole cycles,

N elf. Figure 4 illustrates in two dimensions the physical situation of three

NAVSTAR satellites following each other around the Earth in their 12-hour

period circular orbits. The observer is located at Rand the satellites are+ + +

located at PsI' Ps 2 and Ps 3' The NAVSTAR satellites each transmit

the same frequency (fT), and the observer is assumed to have a receiver with

an oscillator synchronized to the same UTC frequency scale. The measured

Doppler frequency is

(6)

The velocity vector projections along the line of sight cause the Doppler

shift effect. Consider a simplification of a non-rotating Earth or a possible

observer at a very high latitude so that the Earth's rotational velocity is

very small compared to the satellite's velocity along its orbital path. Each

value of the Doppler frequency, fM, measured at time t, allows the look-up

or computation from ephemeris elements of the satellite state vector, contain­

ing the position P. and velocity V .• Therefore, all possible placesS1 S1

where the observer could be located are on the surface of a cone whose vertex+is at the satellite position p , and whose axis of rotation lies along the

ssatellite's velocity vector. The half apex angle of the cone is given by

(7)

Since Figure 4 illustrates a two-dimensional situation, the intersection of

the cones with the plane of the figure reduces the loci of possible positions

13

,/,/

,//'

/',/

,/

Figure 4. SERIES Doppler positioning ~eometry

,

to lines of constant Doppler. The intersection of the lines of possible

position from satellites 1 and 3 are sufficient to determine the observer

position R. Satellite 2 is shown at the special case of zero Doppler occurring

at the time of closest approach (TCA) so that the observer must lie in a plane

defined by the normal to the satellite 2 velocity vector. Actually, two satel­

lites would have been sufficient to determine R given the assumption of

synchronized receiver and satellite oscillators, but the third satellite adds

redundancy or an additional observation to solve for the receiver reference

oscillator frequency offset relative to the assumed synchronized satellite

transmitted frequencies fT.

Having

P'D =1 Ps

+ -I-determined R and having Ps as a given, the Doppler pseudo range

RI is deduced and equated to Equations (3) and (4). Thus,

N clf (f T)c/fm

(8)

is a solution for determining the ambiguity of the whole cycles, N c/f.

15

IV. IMPLEMENTATION

The innovations to be described were developed primarily in pursuit of

extraterrestrial geodesy by a system called Satellite Emission Range Inferred

Earth Surveying (SERIES) (Reference 8). However, the term SERIES used in these

descriptions will be considered as a generic term. Imagine one or more SERIES

devices receiving signals from four or more GPS satellites. Each station is

equipped with an antenna, a receiver and a data system (see Figure 5). The

challenge is to extract information from an encrypted spread spectrum, which

is bi-phase modulated with full carrier suppression using an unknown pseudo

random noise (PRN) code sequence. It is known (Reference 6) that the carriers

and rate of digital modulation transitions are controlled by the satellite­

borne atomic frequency reference. All GPS satellites have an identical rate

of 10.23 mega-transitions per second which is termed the chip rate of the

P-code channel. Two other modulations are also present: 1.023 MHz, called the

coarse acquisition channel (CiA), and a 50-bit-per-second telemetry channel

containing satellite ephemeris data. Dual frequency bands are provided for

ionospheric calibration, and are L1 , centered at 1575.42 MHz, and L2 , at

1227.6 MHz. These are generated by multiplying the 10.23 MHz signal by 154

and 120, respectively.

To exploit these signals without code sequence knowledge, a nonlinear

detection scheme is used. The P and CiA code modulation transitions are

recovered when a delay and multiply method is employed (Figure 6). The effect

is to collapse a broadband 20.46 MHz signal (10.23 MHz, double sideband) into

a simple sinewave. Such methods have been common practice (References 9 and

10) to demodulate telemetry data synchronously, where the recovery of the

telemetry bit clock frequency is merely a means to the end of recovering the

telemetry bit stream. In the SERIES implementation, the actual P-code and CiA

code bit sequences are not explicitly used. The strategy is to exploit the

chip rate as a type of coherent range code which has ambiguities that can be

resolved. Specifically, by exploiting the fact that Doppler Positioning is

virtually unique, the ambiguities in the phase measurement of the CiA code can

be resolved, which in turn enables the resolution of P-code cycle ambiguities.

16

TIME---~ INTERVALr--_~ COUNTER

L1P-CODECHIP

L1CIA CODECHIP

FLTR1575±IO

SPREAD£RA-10 L

1+10 MHz

L1 1575.42 MHz

B.P.~----t~ FI LTER

1.1 TO1. 7 GHz

L1

&L2 FROM

GPS SATELLITE

~~

L2 1227. 6 MHz

FLTR1227±10 L2

P-CODECHIP

STATION ~ ~

CLOCK

Figure 5. SERIES simplified block diagram

B. P.FLTR

10.23 M b/sOR 1. 023 Mb/s

DELA "1-----,(

/2

49 nsOR 487 ns

rvvv10.23 MHzOR1.023 MHzSINEWAVE

I-t-T..j

UU-

-

~: - ......, II I

---: ~T/2I I

INCOMINGSIGNAL: s(t)

DELAYED +1SIGNAL: s(t-T/2)

-1L.L.JCl

+1::::>m(t) = s (t)I-

-J x s(t-T/2)a..-1..... ~

cc <C

BIT CLOCK +1SIGNAL, c,c + n = m(t) 0

TRI-LEVEL +1PSEUDO-RANDOM 0NOISE VARIABLE, n -1

TIME

Figure 6. SERIES delay and multiply technique

For some applications, there is a desire to have access to the L-band RF

carriers and the P and CiA-code channels' sub-carriers, all of which have been

coherently derived from the on-board spacecraft atomic frequency references.

Because of the needs of the conventional GPS mission, these signals are stable

to better than 10-12

sec. at 100 seconds averaging time (Reference 11).

Using these signal-squaring techniques, with a half-chip time delay in

the case of the sub-carriers, it is possible to recover the P-code modulation

rate of 10.23 MHz at both the L1 and L2 channels. These P-codes exhibit

Doppler shifts that span from 0 to ±27 Hz. The CiA-code modulation of 1.023

~lHz, usually present only on the L1 channel, is similarly recovered and has

a Doppler shift of 0 to ±2.7 Hz.

A major feature of the SERIES technique is now evident, that of infor­

mation compression. In the beginning, there was a spread spectrum of 20.46

MHz width of unknown modulation, and now all the information necessary for

pseudo ranging and ionospheric calibration is contained in sinewaves which are

confined to a relatively narrow bandwidth. Specifically, for an Earth-fixed

observer all the P-code information is contained within a S4 Hz bandwidth, and

the CiA channel within S.4 Hz.

Imagine now a receiver that has input to it a signal from a single GPS

satellite. The CiA channel output of the 1.023 MB/S chipping rate is connected

to the stop channel of a time interval counter (TIC) (Figure 5). This counter

is started by a 1 pps signal from the receiver clock. The range of the satel­

lite to the receiving antenna is now indicated by the time interval counter,

modulo 978 nanoseconds (the period of 1.023 MHz) with an unknown possible bias

between the satellite and the SERIES receiver clocks. A second receiver can

now perform the same measurement relative to its own clock (Figure 7). By

calculating the following

a differential phase delay observation is created where T. is the time1

interval counter value of the i th station, and the product T.f. is the1 1

fractional phase measured at that station. This is identical to the delay

19

WHERE

4>2 4>1 1 APDELAY, 1" • - - - + -(N - N )=-

f2 f1

f 2 1 c

'2 2(Po -p)-B

8 = 2 12Pl

1 A"::' 61" • C (-s· B) + 1"( + 1"t - C

1"

(N2 - Nl ) RESOLVED FROM DOPPLER DATA

BASELINEVECTOR &INSTRUMENTBlAS ESTIMATION

DELAYOBSERVABLEGENERATION

No

SATELLITEEPHEMERIDES·

->. ->. ~

PI' P'! s CALI BRATION, 1"

Figure 7. A schematic representation of the SERIES baseline solution

data type of VLBI (Reference 12) for which considerable baseline measurement

experience exists (Reference 13). Note that the possible CiA-code phase bias

in the satellite now vanishes, but in its place another bias appears which is

the difference of clock epochs between station 1 and 2. The delay observable

created by differenced CiA-code ranging is ambiguous at the CiA-code

wavelength of 293 meters. Similarly, this same procedure could be employed

using the P-code which would introduce ambiguities in the delay of 29.3m. For

convenience and reduced cost of implementation, the recovered sinewaves are

coherently down-converted to lower frequencies where the performance of simple

time interval measurement equipment is quite adequate (Figure 5).

The next step is to estimate the baseline vector and deduce the time

offset of the station 1 clock relative to the station 2 clock. The conceptual

approach is identical to that employed in quasar VLBI geodesy (Reference 13).

Consider four or more CPS satellites above the horizon common to stations 1

and 2. Four delay observations are developed by differencing digital time

interval counter readings scaled by the Doppler frequency. All of the CPS

satellites are widely separated in the sky, and thus the geometry and the four

delays allow for a simultaneous estimation of the three-dimensional baseline

vector and the relative clock bias between the stations (X, Y, Z, ~t). The

strategy here is to make four delay observations as rapidly as possible so as

to minimize the effect of receiver clock instabilities. The clock error

contribution grows linearly with the total time required to observe the four

or more CPS satellites.

21

V. DIFFEkENTIAL OBSERVATIONS

An important data type is differential Doppler and differential pseudo

ranging extracted either from the received carriers or from the P or CiA

channels. Differencing automatically removes any possible phase or phase rate

biases which might originate in the GPS satellites t because a bias which may

be unique to one satellite is simultaneously recorded by each user station and

is subsequently cancelled in the difference. Operating in a differential mode

also reduces the sensitivity of the baseline vector determination to the

uncertainties within the satellite ephemeris. Orbital errors contribute to

the baseline error and are attenuated by the ratio of the baseline length to

satellite height t nominally 20 t OOO km (Reference 14). Thus t for a 100 km

baseline separation between receivers t a 10 meter satellite orbit uncertainty

at its 20 t OOO km orbital position can contribute t with worst case geometrYt

~ cm uncertainty to the baseline determination.

22

VI. IONOSPHERIC CALIBRATION: DUAL P-CODE CHANNELS

The L-band region of the electromagnetic spectrum contains substantial

ionospheric delay effects when considering the sub-decimeter baseline accuracy

requirements of geodesy. In an uncalibrated mode, the L1 channel at 1575

MHz would experience approximately 15m of zenith range delay or up to 40m of

delay at low elevation angles during active ionospheric conditions (zenith

columnar contents of 1018

electrons per m2). In an uncalibrated Doppler

positioning mode, using the SERIES method, the L1 channel would be disturbed

by approximately 10 meters of positioning-equivalent error which results from17 2a columnar electron content rate of change of 6x10 el/m per hour.

Clearly, the dual channel availability of NAVSTAR must be used to

explicitly measure the ionospheric delay and delay rate effects along the line

of sight to the satellite being observed. The differential, L1 - L2frequencies, while not being optimum for measuring the ionospheric delay, have

sufficient separation for determining the dispersion effect with the required

precision.

In an earlier form of experimentation (Reference 15) the SERIES

ionospheric calibration used an analog cross-correlation approach which

exploited the fact that the identical P-code modulation sequence is coherently

transmitted by the NAVSTAR satellites on both the L1 and L2 GPS channels

(Reference 6).

The ionospheric columnar content was measured by intentionally delaying

the first arriving L1 signals and then cross-correlating with the L2channel. The delay in the L1 is made adjustable to compensate for the

differential delay experienced in the L2 channel. When the output of the

correlator is a maximum, the variable delay value is the ionospheric delay.

While the technique was functional, it suffered from two difficulties: The

measurement was difficult to calibrate to 0.1 nanoseconds and it was vulnerable

to changes in the NAVSTAR transmission format. For example, the GPS satellites

could conceivably be reconfigured so as to broadcast different portions of the

P-code sequence on L1 and L2 thus destroying the possibility of obtaining

an output by inter-channel cross correlation.

23

A more reliable approach was to exploit the 10.23 MHz P-code bit transi­

tion rates separately derived from the L1 and L2 channels. The 10.23

megabit-per-second (MBjS) code chip rate is directly derived from the on-board

atomic frequency references and used to coherently modulate L1 and L2•

Although the actual code sequence or code phasing might be altered, it is

likely that the P-code chip rates of L1 and L2 will not be changed. Thus,

the phase of the code transitions are probably in coincidence or with simple

constant offsets in the satellites and will therefore experience ionospheric

dispersion in the same manner that the code sequence is differentially delayed

in the previously mentioned inter-channel correlation approach.

A digital time interval counter and/or a phase measuring device is used

to determine the lag of the LZ channel relative to the L1 channel to deter­

mine the columnar content of the ionosphere. This SERIES subsystem has been

given the name SLIC for Satellite L-band Ionospheric Calibration (Figure 8),

and is a potentially powerful tool for mapping ionospheric conditions.

Consider the P-code transitions at L1 and L2 to be phase coincident

at the satellite's location. The L1 and L2 channels then experience the

ionospheric dispersion because of their 347.82 MHz of frequency separation.

The- physics of this situation are discussed in Reference 16 and summarized as

follows: The group delay effect is in general, t = AI/cf 2 , whereg

A = 40.3 in MKS units

I columnar electron content, electrons per 2= m

f radio frequency, Hertz

c speed of light, meters per second

t = ionospheric time delay, secondsg

The SLIC observable is the differential group delay (Tg ) between L1 and

L2:

Consider now the extreme conditions of the ionosphere. The mlnlmum

zenith content to be expected is I = 0.5x10 17 electrons/meter 2 which will

24

DIRECT IONOSPHERICDELAY MEA SUREMENT

START,-------,STOP TIME INTERVAL,.--~ COUNTER

J...

L1 AT

1575 MHz

L2 AT

1227 MHz

RFAMPL

NV1

Figure 8. SERIES technique for measurement of the columnar chargedparticle content

result in r (min) = 1.7~ nanoseconds.g Its L

could approach 10 el/m and observingIts 2

columnar content of L.~xl0 el/m, and

seconds. Exploiting the P-code in

At a maximum, the zenith content

at a low elevation angle causes a

results in T (max) = 87.5 nano-g

this manner has an ambiguity of ~7.75 ns,

between L1 and L:2. (267

on the L 1 channel).

observable at either L 1 or

'r'h 1017 1/· :2. l'us, e m resu ts In

and thus care must be exercised to avoid ambiguous measurements in high

ionospheric content situations.

In geodetic missions, calibration must be applied to the pseudo-range

L2• The ionospheric delay on L1 is t g1 = 1.54 Tg

•

T 3.5 ns or 105 em path equivalent differenceg

em path delay on the L2 channel minus 162 em path delay

The SLIC error sources contain three major elements: (1) receiver noise,

(L) receiver delay, and (3) possible biases in the NAVSTAR at the time of

transmission. The measures taken to compensate for these error sources are as

follows:

(1) Using a 1.5 m diameter antenna provides more than adequate signal

strength, and the differential phase shift between the L1 and

L2 channels is measured to within 0.1 ns RMS or better with one

second of averaging using the Bipolar transistor preamplifier,

corresponding to a columnar content of 3x1015 el/m2 or 4.6 em

of pseudo range calibration on the L1 channel. Thus, to achieve

a two centimeter precision in pseudo ranging calibration, 5 seconds

of SLIC data must be averaged.

(L) To measure the differential phase behavior of the SLIC receiver

itself, a GPS satellite simulator consisting of a PRN (pseudo-random

noise) sequence is modulated onto carriers at L 1 and L2 and

injected into the first RF amplifier, and the data system processes

the signal in a manner identical to the reception of the NAVSTAR

signals. In this way the dissimilarities between L1 and L2signal paths are measured with a precision of 0.06 ns to achieve a

2. em accurcy in the SLIC receiver delay.

26

(3) According to the specification of the Block I NAVSTAR satellite

(Reference 17), the relative phasing of the P-code transmissions on

L1 and L2 channels will vary by less than 1.5 nanoseconds.

Such a time difference in transmission will cause a bias in the

deduced total columnar content equivalent to 3xl016el/m2

but

will have no effect upon the SERIES determined baseline vectors

performed in a two-station differenced pseudo range mode since any

satellite biases are common-mode rejected.

Important features of the SLIC method are listed below:

1) Total columnar electron content measurement with a precision of

10l5el/m2 with 10 seconds of averaging and an accuracy dependent

upon phasing in the NAVSTAR, limited to approximately 3xl016

el/m2

•

2) No P-code digital sequence knowledge required.

3) No dependence upon orbit kinematics of the NAVSTAR.

4) No modeling of geomagnetic fields or unsensed electrons as with the

Faraday Rotation method.

5) No active elements as with ionosondes or topside sounders aboard

Earth-orbiting spacecraft.

6) Modest instrumentation requirements, with options for omni­

directional reception.

7) Columnar content measurement from a single station; no cross­

correlation between two stations to achieve signal detection as in

the VLBI method.

8) A single satellite pass allows horizon-to-horizon sampling of the

ionosphere in 6 hours; the full 18 satellite constellation will

<lllow a continuous sampling in at least four directions (perhaps as

many as seven) from every location on the Earth's surface.

27

VII. BASELINE ESTIMATION SOFTWARE

Baseline estimation software has been written to allow the evaluation of

the SERIES receiver engineering tests. This program allows the determi­

nation of a baseline between two SERIES stations, the clock offset between the

stations, and the approximate location, in the WCS-72 frame, of each station.

Data input from the SERIES stations to the baseline estimation soft­

ware consists of time tagged pseudo ranges to the CPS satellites as well as

measurements of the ionospheric delays along the lines of sight to the CPS

satellites. Other ancillary data can be entered such as surface measurements

of barometric pressure, temperature, relative humidity, or water vapor

radiometer measurements of water vapor delays.

The baseline program also requires ephemeris information. The orbits

determined post facto by the Naval Surface Weapons Center (NSWC) are currently

being used for SERIES baseline tests. These orbits, with approximately 10-m

errors, are adequate for precision determinations of baselines up to 100 km.

The baseline program operates on singly differenced ranges, i.e., the

difference in simultaneous pseudo ranges between two stations to a CPS

satellite. These differences are formed and compared with the modeled

differences based on the ephemerides, the ~ priori station locations and clock

offsets. The three components of the baseline vector and the clock offsets

are adjusted so that the least squares residuals are obtained.

The data can be sent over phone lines so that a near-real time (within I

hour) baseline solution can be obtained. Presently, this near-real time

solution does not have the precision ephemeris information available, leading

to degraded baseline accuracy. The ~ priori ephemeris is available at the

time a baseline is measured, while the post-fit ephemeris is generally avail­

able two weeks hence.

28

VIII. SERIES TEST SEQUENCE

The testing program was designed to allow a sequential check of suspected

system error sources (see the listing below).

Experiment

Single Station

o Baseline

l~U M Haseline

150 M Cornman Clock

I~U M Attenuated

l~O M Long Track

Ll KM (DSSIJ/14)

II KM (Clock Sync)

Faraday Co-Location

171 KM (JPL/Goldstone)

A. SINGLE STATION TESTS

System Component Tested

Instrumentation

Instrumentation + Baseline Estimator

Instrumentation + Clock + SNR + Multipath

Isolates Clock Error Source

Isolates SNR Error Source Signal

Sensitive to Multipath

Test Field Operation of Mobile Units

SERIES Clock Sync Demonstrated

SERIES Ionospheric Calibration Validated

Tests Dependence of Errors on Baseline Length,

Atmospheric Delay Models, Ephemeris Errors

Single station tests were performed to indicate whether receiver

stability was adequate to obtain baseline measurement results with few cm

precision. The SNR (Signal to Noise Ratio) of the GPS satellite signals in

the receiver Intermediate Frequency (IF) band were also measured in single

station tests. This was compared with the output SNR of the reconstructed

code clocks in order to verify that the SERIES code clock recovery technique

was operating at or near the theoretically predicted performance level.

Single station tests were repeated throughout the test sequence in order

to periodically validate instrument performance.

H. 0 BASELINE

The 0 baseline tests were performed with the output signal from a single

antenna being split to provide inputs to the two SERIES receivers. One

29

Rubidium frequency standard was used to provide the reference clock for both

of the SERIES receivers. The RMS scatter in the baseline lengths was 1 cm,

which is less than the J cm predicted by the SNR if independent noise is present

in each receiver. This is due to the fact that much of the noise in the abaseline experiment is correlated between the two receivers and, therefore,

cancels. The mean measured baseline length of 5 cm is thought to have peen

caused by electronic cross-talk between the two closely situated receivers.

c. 150 METER BASELINE

As can be seen in Figure 9, a greater separation to a 150m baseline

increased the RMS scatter to 3cm, which is the value expected from the SNR

error combined with the clock jitter error resulting from the Rubidium vapor

frequency standards. The mean of the SERIES length measurements differed from

a conventional survey by O.3cm.

The experiments conducted with a common frequency standard for the two

stations showed no significant improvement over those which used separate

clocks. This is consistent with the expected l-cm baseline error contribution

due to clock noise.

A baseline measurement was conducted with a known attenuation added to the

satellite signal input to the first amplifier. This caused a known degradation

ot the system SNk. From this information, the increase of that part of the R}IS

scatter of delay residuals in the baseline fit due to system SNR could be

predicted. By noting the actual increase in the RMS of the delay residuals, it

was determined that, in fact, the system noise from the first amplifier was the

dominant error source in the SERIES GPS receivers.

lhe errors due to the presence of multipath signals are expected to be

small for a GPS receiver with a highly directive antenna. These errors are

ditficult to estimate, and so were measured by having both stations track a

common satellite while subjecting one station to likely multipath sources. In

this experiment media delays, geometric effects, etc., were largely common­

mode rejected. The multipath effect was determined to cause less than a I-em

baseline error, using tile SERIES 1.5-m antenna.

30

151. 65

t-V)

0:::151. 60LU

t-LU

~:::r::

0t-<:;)z

151.55LU-I

LU dL CALCULATEDz SURVEY-I

00FROM SNRLU

v..>V)

UMEAN)...... «ca 151. 50

151.45 ---6""'-/5-~--I.....----I-2"'/6-..Io.--4""'-/-5"""/ --------'"13/5 1/6 3/6 6 6

EXPERIMENT DATA (DAY/MONTH 1982)

Figure 9. SERIES 150 m baseline length measurements. Each circle withouterror bars represents a solution using 1.25 to 1.75 hours ofdata taken on a constellation of five GPS satellites, mutuallyvisible over Southern California. The scatter among thesesolutions is consistent with the SNR error bar. The meanlength agreed with the conventional survey to within theSERIES errors, with a difference of 0.3 em.

D. 21 KM BASELINE

For the proof-of-concept system~ the SNR errors still dominate at the

21km baseline~ with 4cm RMS scatter of the solutions for baseline length. The

comparison between the composite survey (combined VLBI and conventional

measurements) and the SERIES measurement~ in Earth fixed coordinates~ is shown

in Figure 10. The errors in the composite survey are estimated to be less

than 5 cm in each component. The errors in the components of the SERIES

determination vary. The north-south component of the SERIES baseline is quite

well determined~ while the east-west and vertical components are less well

determined. This can be explained by simple geometric factors. The set of

five operating GPS satellites available during the tests described here formed

a constellation over California which is nearly aligned along the north-south

direction. This geometry provides for a weak solution in the east-west

component of the baseline. A similar argument can be made for the local

vertical component~ as obviously no satellites can be observed in the local

"down" direction. This effect is~ of course~ also present for the other

baselines which were measured. The existence of the eventual full 18

satellite GPS system in the late 1980's will offer improved geometric as well

as time coverage.

E. 21 KM - CLOCK SYNCHRONIZATION

While the SERIES stations were on the 21 km baseline~ two experiments

were performed to demonstrate the measurement of the clock offset between the

two station clocks (Ref. 5). The usual SERIES multi-parameter solution

contains a determination of clock offset and rate as well as the baseline

vector separating the two stations. This solved-for clock offset contains

contributions from the differential receiver delays at the two SERIES

stations. In addition, the solved-for delay includes contributions from the

phases of locally generated heterodyne signals. Ancillary measurements were

taken during the clock sync experiments to calibrate the effect of each SERIES

station's heterodyne signal phase. In order to measure the differential

instrumental delays between the two receiving stations, they were brought

together for a calibration run on a 2m baseline. The error in the SERIES

measurement of clock sync was estimated to be 1 ns (10-9sec). This is

32

*COMPOSITE SER IES SER IES -

COORDINATE SURVEY (ERROR) COMPOSITE

LENGTH 21463.655 21463.667 +0.012(METERS) (±0.018 )

X -2531. 786 -2531.829 -0.043(METERS) (±0.021 )

Y 14090.312 14090.266 -0.046

(METERS) (±0.027 )

Z 15991. 9.25 15991.960 +0.035(METERS) (±0.021 )

* SURVEY IS COMPOSITE OF VLBI + CONVENTIONAL SURVEY MEASUREMENTS

Figure 10. Comparison of SERIES and composite measurements of the21 km baseline

dominated by errors in measuring the differential instrumental delays between

the two receivers. On longer baselines, the ephemeris errors would also become

important. They would be expected to contribute a 5 ns error to the clock sync

measurements on a 3,000 km baseline if 10-meter orbit errors are assumed.

During the two SERIES clock sync experiments, a Rubidium vapor frequency

standard was used as a traveling clock to allow an independent check on the

SERIES clock sync results. To obtain accurate results with the traveling

clock, three round trips were made between the two SERIES stations. At each

visit of the traveling clock to a SERIES station, the epoch of the traveling

clock was compared to the SERIES station clock using a time interval counter

to measure the offset of the clocks. Care was taken to insure that trigger

levels were set to the same levels on each visit and the same coaxial cables

were used at each station. The offset between each station clock and the

traveling clock was later fit to a second order polynomial so that the time

offset between the two station clocks could be determined at a common epoch.

The traveling measurement clock error was estimated to be 0.7 ns.

When the clock offsets obtained with the SERIES technique were compared

with the independent traveling clock measurements, interpolated to the same

epoch, agreement was within one nanosecond for each of the two clock sync

experiments (Fig. 11). (This agreement was obtained after adjusting the 977-ns

ambiguity of the CiA code chip period present in the SERIES receiver.)

F. FARADAY CO-LOCATION - SLIC TEST

A schematic representation of the operating principle for the SERIES

L-band Ionospheric Calibration (SLIC) technique is shown in Fig. 12. In order

to demonstrate this technique for measuring the columnar free electron content

in the ionosphere, data were taken with the SERIES station co-located with

another system capable of measuring the ionospheric content (Faraday Rotation).

Figure 13 shows a comparison of the two techniques for one day of data. Note

that each technique measures the total electron content along different lines

of sight, and these are then mapped to the zenith content. The agreement

between the two techniques is about 10 16 e/m2 when the two lines of sight

are close and, therefore, mapping errors are small.

34

tSTA A - tSTA B t STA A - tSTA B DIFFERENCESER IES RECE IVER TRAVELING CLOCK BETWEEN 2 TECHN IQUES

23 AUG 82 -189.7 ns -190.3 ns +0.6 ns14:36:45 PDT

24 AUG 82 -1302.0 ns -1301.4 ns -0.6 ns14:00:45 PDT

Figure 11. One nanosecond (10- 9 second) agreement was observed on each oftwo experiments in which a traveling clock was used to verifySERIES clock synchronization measurements made on the 21 kmbaseline.

ICDELAY MEASUREMENT

L1 AT1575 MHz

START _r-- I-

RF --. TIME INTERVALAMPL STOP _ COUNTER--.

L.-- I- L2 AT" T ./

1227 MHz DIRECT IONOSPHER

FARADAY SATELLITE

~GPS~ SATELLITE

L1 P-CODE L2

P-CODE

ION

0

S~~'~:~;;~;li\\,\'\\\'\\\\\\'\\\'\\\\\\\\\'\\'\\\\\!\\\;'\\\:\\~!'~j\t\\\;\'i\;\:\\\\\\\\i\\\\\\\\\\\\\'\\\\\\\\\\\·\:\\\\\\\\\\\;;;\\\t:;;,

Figure 12. The P-codes are nearly in phase when transmitted from the GPSsatellites, but suffer a differential delay due to the presenceof free electrons in the ionosphere. By measuring the magnitudeof this differential delay, the SERIES station is able todetermine the columnar content of free electrons in the ionosphere.

5.06- CO-LOCATED FARADAY ROTATION MEASUREMENT

-------Q)I~ 0 Sli CMEASUREMENT~r- 4.0.......~0

~

1-'ZLJ.JI-

3.0z0 CLOSEST APPROACH OF FARADAYuz AND SliC lINE-OF-SIGHTS ~

00::: 9I-U 2.0LJ.J-oJLJ.J

w -oJ-...J «

I-0

{ 5.4 ns L1 DELAYI-

~ 1.0t= 8.9 ns L2 DELAYzLJ.JN

004:00 07:00 10:00

1 DEC 1982TIME OF DAY (PST)

Figure 13. The SERIES ionospheric measurements are shown as circles, whilethe Faraday rotation results are triangles. The rapidly increas­ing TEe is due to the fact that the data were taken during sun­rise. It can be seen that 'the agreement is better when the 1inesof sight for the two techniques are close, as this allows mappingerrors to cancel.

G. 171 KM BASELINE

The 171 km baseline tests are sensitive to ephemeris errors. (Fig. 14).

Tropospheric delay corrections were applied to the baseline solution based on

surface weather data (temperatures pressure and relative humidity) taken during

the experiments. Ionospheric delay or SLIC corrections have been derived from

the dual band SERIES data and applied. On the 171 km baseline between JPL and

Goldstone, the error contribution due to a 10-meter uncertainty in the satel­

lite ephemerides is estimated to be 9 em.

The KMS scatter of the baseline lengths was 7 em for the 171 km base­

line. This is consistent with the presence of 10-meter errors in the

ephemerides supplied by the NSWC.

H. BASELINE SUMMARY

A summary of the SERIES baseline testing results is given in Fig 15.

This table represents the precision of the SERIES baseline measurement by the

RMS scatter of the individual day's solutions. Typically, there were 12 days

of data for each basel~ne. The accuracies of the SERIES measurements are

shown through comparison with independent surveys.

I. BASELINE ERROR VS BASELINE LENGTH

Figure 14 shows the calculated effect of various error sources on baseline

length determination accuracy for the current SERIES proof-of-concept system vs

baseline length. For the shorter baselines, the SNR is the dominating error

source. The errors due to both the SNR and clock terms can be reduced to less

than I-em by more efficient data collection combined with more nearly simul­

taneous observations of the set of GPS satellites. The error shown for water

vapor is based on the use of surface data to correct delays, and could be

reduced with the use of current water vapor radiometers (WVR's). The ephemeris

errors dominate for long baselines. The approach proposed by the JPL GPS

development team to reduce ephemeris errors is to use a set of sites with

positions determined by VLBI as fiducial points for a network of GPS receiver

stations. Those observations made from the fiducial points would be used to

obtain accurate ephemerides of the GPS satellites during each experiment.

38

SERIES ERROR SOURCES fOR PROOf-Of-CONCEPT STATIONS30

• MEASURED RMS IN LENGTH

(1) SURFACE MODEL(2) POST FIT EPHEMERIDES SUPPLIED BY NSWC (± 10 M ACCURACY ASSUMED)

Figure 14. The calculated cumulative error in baseline length is shown forthe SERIES receivers as tested. It can be seen that for base­lines longer than a few tens of km, the error is dominated byephemeris and tropospheric calibration errors. The performanceof the SERIES system is shown by the triangles for the 0,150 m, 21 km, and 171 km baselines.

BASELI NE LENGTH RMS SCATTER IN LENGTHDIFFERENCE OF MEANLENGTH FROM SURVEY

o_ __.._ _. .._.. . ..__1 CM .._ _. __.. .__ _* +5 CM

150 M._..__ ....__ .._. ...._. . . 3 CM. ...._. __...__.... .._... .+0.3 CM

21 KM __.._. .... ....__ ._..... 4 CM . ... ._. .._..__. ...+ 1. 2 eM

171 KM__.._._... ...._. ._....__ 7CM .. ._.__...._._._.**+ 7 CM, -1 CM

*ELECTRONIC CROSSTALK WAS THOUGHT TO CAUSE LENGTH BIAS

** DIFFERENCE FROM ARIES VLBI AND PRELIMINARY TEXAS LASERRESULTS RESPECTIVELY

Figure 15. Table of SERIES baseline results. Between 9 and 12 measurementswere made at each of the baselines.

IX. CURRENT SERIES STATUS

the results of the tests described in the previous section show that the

dominant error sources included primarily system noise and clock jitter on the

short baselines (see Fig. 14). Un the long baselines, orbit errors and the

delays due to the presence of water vapor become important. Since the delays

due to the ionosphere are measured in the SERIES receiver, these delays are

calibrated.

As described previously, the errors due to water vapor delays can be

reduced by the use of water vapor radiometers which measure those delays, and

a network of high precision GPS receivers operating at known sites (fiducial

stations) can be used to greatly reduce the orbital errors.

Improvements have already been made to the SERIES hardware and observa­

tion strategy, since taking the data contained in this report, to reduce the

errors due to system noise and clock jitter. The clock errors were reduced by

changing the time between satellite observations from two minutes to one

minute. Three approaches have been used to reduce the error due to system

noise.

The first of these was the reduction of the bandpass width of the P-code

tracKing filter. This allowed the data sampling to be closer to the desired

Nyquist rate. The SERIES data sampling rate is currently limited by hardware

constraints to ten samples per second. The bandwidth of the sampled signal

was originally bO hz, and therefore had a Nyquist rate of 160 samples per

second. The bandwidth is currently 20 Hz, with a Nyquist rate of 40 samples

per second. The loss of SNR due to undersampling the P-code signal has been

reduced to 0.25 from 0.0625.

The second approach was to replace the Bipolar Transistor preamplifier

with a Gallium ArSenide Field Effect Transistor (GASFET) preamplifier. The

table below shows the SERIES system temperatures, including the antenna

temperature, for either preamplifier, at each of the two frequencies received.

41

Preamplifier

Bipolar Transistor

GASFET

L2 at

1227.6 MHZ

485 K

95 K

L 1 at

1575.42 MHZ

560 K

125 K

The third approach to reduce the effect of system noise on pseudo

range measurement accuracy was to acquire carrier range data. The

capability to obtain carrier data from the L2 frequency band has been

added to the SERIES system. The advantage of carrier range data over

P-code ranges is due to the fact that the recovered (squared) carrier

wavelength iS t for the L2 carrier t 120 times shorter than the P-code

"wavelength t " and the error due to system noise is proportional to the

wavelength. Therefore t for a given SNR on the L2 band t the carrier

range error due to system noise is 120 times less than the P-code range

error due to system noise.

A potential problem with the use of the carrier range data type is

the necessity to determine the pseudo-range ambiguitYt equal to the

wavelength of the squared carrier (12.1 cm for L2). The capability to

determine this ambiguity using the P-code pseudo-range measurement is

predicted from current P-code performance t but has not t as yet t been

tested experimentally.

Due to the high SNR resulting from the SERIES high gain antenna t the

addition of the carrier data type to the already precise P-code data will

make the SERIES receivers a unique tool. Potential uses t in addition to

geodesYt are investigations of inherent limiting error sources of the

space segment of the GPS t such as the possible presence of multipathing in

the GPS satellite antennas. With or without the addition of the carrier

range data type t the SERIES geodetic instrument has been demonstrated to

be a GPS receiver capable of making accurate t cost effective measurements

of baselines from sites located in remote areas.

42

itu+{j

sP

P'

!J.p

!J. P'

!J.t (t u - t s)

!J.t u • !J.t s

!J.t 8i0

S

B

t TMc

f

fIn

(Mu l - MuZ)

N·l

V sN

clf l-

f.l

'1'i

8. f .T.l l. l

f1'1

P'D

t g

1

x. SYMBOL GLOSSARY

User/observer geocentric vector

Satellite geocentric vector

Range from observer to satellite

Measured pseudo range;p' = p + c (!J.tu - !J.t s ) + noise

Differential pseudo range

Measured differential pseudo range

Apparent del~y in arrival of satellite clockpulses

User or satellite clock offset with respectto UTC (Universal Time Coordinated)

Clock offset of i th satellite

Correction term for converting spherical toplanar wave fronts

Unit vector from primary station to satel­lite signal source at distance Pi

Baseline vector from remote station toprimary statio-n

Time delay through transmission media

Speed of light

Code frequency in satellite reference frame

Code frequency measured by the user

Clock offset of clock 2 relative to clock 1

P code or CiA code c~hle ambiguityencountered at the i station

Satellite velocity

Integer number of whole wavelength cycles

Effective wavelength of code at frequency f

F d. th .requency measure at l statlon

Time of reception tagged by the TIC (TimeInterval Counter)

Fractional phase of the CiA or P codes

Measured Doppler frequency

Measured Doppler pseudo range

Group delay effect; ionospheric time delayin seconds

Columnar electron content

C/A code

P-code

Differential ionospheric group delay betweenL1 and L2 channels

Coarse acquisition code with 1.023 MHz bitclocking rate, normally present in L1 CPSband

Precision code with 10.23 MHz bit clockingrate, present in L1 and L2 CPS bands

CPS frequency band centered at 1.57542 CHz

CPS frequency band centered at 1.2276 CHz

44

REFERENCES

1. MacDoran t P. F' t Spitzmesser t D. J' t Buennage1 t L. A' t "SERIES:Satellite Emission Range Inferred Earth Surveying," Proceedings of theThird International Geodetic Symposium on Satellite Positioning t LasCruces t New Mexico t Feb' t 1982.

2. MacDoran t P. F' t "Satellite Emission Radio Interferometric EarthSurveying t SERIES-GPS Geodetic System t " Bulletin Geodesique t 53 (1979)tpp. 117-138.

3. Young t L. E' t Spitzmesser t D. Jot Buennagel t L. Aot "SERIES t A Novel Useof GPS Satellites for Positioning t " Proceedings of the IAG/IUGG Symposiumon Point Positioning in Marine GeodesYt Maracaibo t Feb' t 1983.

4. Royden t R. N' t Miller t R. B' t Buennagel t L. B' t "Comparison of NAVSTARSatellite L-band Ionospheric Calibrations with Faraday-RotationMeasurements t " to be published in Radio Science t May-June t 1984.

5. Buennagel t L. A' t Spitzmesser t D. J' t and Young t L. E' t Proceedings ofthe Fourteenth Annual Precise Time and Time Interval (PTTI) applicationand Planning Meeting t Dec' t 1982.

6. Spilker t J. J' t Jr' t "GPS Signal Structure and Performance Character­istics t" Navigation Vol. 25 t 121-146 (1978).

7. Milliken t R. J' t and Zoller t C. J' t "Principle of Operation of NAVSTARand System Characteristics t " Navigation VoL 2S t Summer t 1978

8. Ong t K. M' t MacDoran t P. F' t "Quasars and Artificial Satellites forGeodynamics Monitoring and Earthquake Research," International Conferenceon Cybernetics and SocietYt October t 1979.

9. Spiker t J. J' t Jr' t Digital Communications by Satellite t Prentice-RalltInc' t Englewood Cliffs t New JerseYt 1977.

10. Viterbi t A. J' t Principles ~ Coherent Communications t McGraw-Rill t 1966.

11. Bartho1omew t C. A' t "Satellite Frequency Standards t " Navigation tVol. 25 t Summer t 1978.

12. Thomas t J. B' t "An Analysis of Long Baseline Radio InterferometrYt" TheDeep Space Network Progress Report t Tech. Rep. 32-l526 t Vol. 7 t p. 37,Jet Propulsion LaboratorYt Pasadena t Ca1ifornia t 1972a.

13. Niel1 t A. E' t K. M. Ong t p. F. MacDoran t G. M. Resch, D. W. Fite,L. J. Skjerve t D. J. Spitzmesser t D. D. Morabito, L. Tanida tE. S. Claflin, B. B. Johnson t M. G. Newsted, A. Banisch t andJ. F. Dracup: "Comparison of a Radio Interferometric DifferentialBaseline Measurement with Conventional GeodesYt" Tectonophysics, 52,p. 532 t 1979.

45

14. Russell, S. S., Schaibly, J. H., "Control Segment and User Performance,"Navigation, Vol. 25, Summer, 1978.

15. MacDoran, P. F., SERIES-Satellite Emission Radio Interferometric EarthSurveying, Third Annual NASA Geodynamics Program Review, Crustal DynamicsProject Geodynamics Research, NASA, Washington, D.C., pp. 76 (1981).

16. Koehler, R. L., "Radio Propagation Measurements of Pulsed Plasma StreamsFrom the Sun Using Pioneer Spacecraft," J. Geophysical Research 73, 115,4483-4894 (1968).

17. Interface Control Document, 03953, Space Division, Rockwell InternationalCorporation, September, 1975.

46 NASA-JPL-Coml.• L.A., Cain.